蜗轮减速器振动信号特征提取与状态识别

摘要/Abstract

摘要：

为实现蜗轮减速器运行状态识别，首先结合小波包分解和矩阵理论的特点，提出基于参考信号的小波包能量矩阵构造方法，分析了矩阵的最大奇异值（特征值）与运行状态的物理联系，并验证了所提方法比以往方法提取出的特征参数敏感度更高;然后改进思维进化算法（MEA）用于优化BP神经网络，实现对运行状态的智能识别，将提取的特征参数构成神经网络的输入向量，结果表明识别正确率提高了17.93%，从而验证了改进算法的优越性;最后提出了一种快速分类方法，该方法可以较好地区分故障与正常状态，解决了对实时性要求较高的在线诊断问题。

关键词: 相对能量矩阵, 特征提取, 思维进化算法, 识别

Abstract:

To achieve state recognition of worm reducer running state, a method of constructing wavelet packet energy matrix was proposed firstly based on reference signal. This was done by combining the characteristics of wavelet packet decomposition and matrix theory. It analyzed the intrinsic physical relationship between the maximum singular value (eigenvalue) and operating conditions. The results validate that the feature parameters extracted by this method is more sensitive than that by conventional ways. Secondly, the MEA was improved to optimize BP neural network. The fault characteristic parameters were extracted as the input feature vectors of neural network to realize the recognition of worm gear reducer states. The experimental results show that correct diagnosis rate increases 17.93 percent, which indicates the superiority of improved algorithm. A fast classification method was presented, which can solve the problems of on-line diagnosis for real-time requirements by distinguishing between normal and failure states well.

Key words: relative energy matrix, feature extraction, mind evolutionary algorithm (MEA), recognition

中图分类号:

TP39

白国振, 周海宁. 蜗轮减速器振动信号特征提取与状态识别[J]. 中国机械工程, 2016, 27(03): 349-354.

Bai Guozhen, Zhou Haining. Vibration Signal Feature Extraction and State Recognition for Worm Reducer[J]. China Mechanical Engineering, 2016, 27(03): 349-354.

参考文献

[1]钟先友，赵春华，田红亮,等. 基于自适应最大相关峭度解卷积和频率切片小波变换的齿轮故障特征提取[J].中国机械工程, 2014, 25(21):2880-2885.

Zhong Xianyou, Zhao Chunhua, Tian Hongliang, et al. Gear Fault Feature Extraction Method Based on Adaptive MCKD and FSW[J]. China Mechanical Engineering, 2014, 25(21):2880-2885.

[2]王冬云,张文志. 基于小波包变换的滚动轴承故障诊断[J]. 中国机械工程,2012,23(3):295-298.

Wang Dongyun, Zhang Wenzhi. Fault Diagnosis Study of Ball Bearing Based on Wavelet Packet Transform[J]. China Mechanical Engineering, 2012, 23(3):295-298.

[3]李小玉，何怡刚，李目，等.基于小波分析和遗传神经网络的模拟电路故障诊断方法[J].计算机应用研究,2011,28(12):4517-4519.

Li Xiaoyu, He Yigang, Li Mu, et al. Wavelet Decomposition and Genetic Algorithm Based Approach for Fault Diagnosis of Analog Circuits[J]. Application Research of Computer, 2011, 28(12):4517-4519.

[4]李国宾，关德林，李廷举. 基于小波包变换和奇异值分解的柴油机振动信号特征提取研究［J］. 振动与冲击，2011,30(8):149-152.

Li Guobin, Guan Delin, Li Tingju. Feature Extraction of Diesel Engine Vibration Signal Based on Wavelet Packet Transform and Singularity Value Decomposition[J]. Journal of Vibration and Shock, 2011, 30(8):149-152.

[5]齐俊德,李山,陈冰. 基于小波包能量矩阵的轴承信号特征提取[J]. 振动与冲击,2013,32(21):107-111.

Qi Junde, Li Shan, Chen Bing.Feature Extraction of Bearing Vibration Signals Based on Wavelet Energy Matrix[J]. Journal of Vibration and Shock, 2013,32(21):107-111.

[6]花汉兵.基于小波包的振动信号去噪应用与研究［J］. 噪声与振动控制，2007(6):19-21. Hua Hanbing. Application and Research of De-noising of Vibration Signal Based on Wavelet Packet[J]. Noise and Vibration Control, 2007(6):19-21.

[7]Shin K,Feraday S A,Harris C J,et al.Optimal Autoregressive Modelling of a Measured Noisy Deterministic Signal Using Singular-value Decomposition[J].Mechanical Systems and Signal Processing 2003,17(2):423-432.

[8]Yang W X,Peter W T.Development of an Advanced Noise Reduction Method for Vibration Analisis Based on Singular Value Decomposition[J]. NDT&E International,2003,36(6):419-432.

[9]孙承意,孙岩,谢克明,等.思维进化——高效率的进化计算方法[C]//第三届全球智能控制与自动化大会论文集.合肥: 中国科学技术大学,2000:118-121.

[10]王小川，李洋，郁磊，等.MATLAB神经网络43个案例分析[M].北京：北京航空航天大学出版社，2013.

[11]刘建霞，李楠，谢克明. 改进的思维进化算法在阵列天线综合中的应用[J]. 测试技术报,2009,23(4):331-336.

Liu Jianxia, Li Nan, Xie Keming. Application of Improved Mind Evolutionary Algorithm to Antenna Arrays Synthesis[J]. Journal of Test and Measurement Technology, 2009,23(4):331-336.

[12]Papadimtrou C H,Steiglitz K.Combinatorial Optimization:Algorithms and Complexity[M].Upper Saddle River,NJ:Prentice-Hall,1982.

[13]赵学智,叶邦彦,陈统坚. 矩阵构造对奇异值分解信号处理效果的影响[J]. 华南理工大学学报(自然科学版),2008,36(9):86-93.

Zhao Xuezhi, Ye Bangyan, Chen Tongjian. Influence of Matrix Creation Way on Signal Processing Effect of Singular Value Decomposition[J]. Journal of South China University of Technology (Natural Science Edition), 2008,36(9):86-93.

[1]	许鹏, 方舟, 王平, 耿明, 许勇. 基于脉冲漏磁暂态特征的缺陷量化评估方法[J]. 中国机械工程, 2021, 32(07): 860-866,881.
[2]	江志农1;王子嘉1;张进杰2;黄翼飞3;茆志伟2. 基于能量算子梯度邻域特征提取的核电应急柴油发电机组故障诊断方法[J]. 中国机械工程, 2021, 32(05): 617-623.
[3]	张琨, 陈定方, 高铭鑫, 黄永亮, 刘洋, 门燕青. [车辆工程与测控]有监督学习下的市域铁路隧道结构裂缝边缘识别方法[J]. 中国机械工程, 2021, 32(04): 446-453.
[4]	李茂月;吕虹毓;王飞;贾冬开. 基于机器视觉的智能车辆鲁棒车道线识别方法[J]. 中国机械工程, 2021, 32(02): 242-251.
[5]	唐贵基;李楠楠;王晓龙. 综合改进奇异谱分解和奇异值分解的齿轮故障特征提取方法[J]. 中国机械工程, 2020, 31(24): 2988-2996.
[6]	万书亭;彭勃;王晓龙. 基于双时域变换和稀疏编码收缩的滚动轴承早期故障诊断方法[J]. 中国机械工程, 2020, 31(23): 2829-2836.
[7]	孙殿柱1;沈江华1;贾宗福1;李延瑞2;林伟1. 散乱点云局部形貌标架量化及特征识别方法[J]. 中国机械工程, 2020, 31(19): 2332-2339.
[8]	蔡艳平;范宇;陈万;张金明. 改进时频分析和特征融合在内燃机故障诊断中的应用[J]. 中国机械工程, 2020, 31(16): 1901-1911.
[9]	何彦1;凌俊杰1;王禹林2;李育锋1;吴鹏程1;肖圳1. 基于长短时记忆卷积神经网络的刀具磨损在线监测模型[J]. 中国机械工程, 2020, 31(16): 1959-1967.
[10]	余石龙, 鲍劲松, 李婕, 张启华. 基于误差注意力的晶圆制造数据异常检测[J]. 中国机械工程, 2020, 31(14): 1686-1692.
[11]	姜少飞;邬天骥;彭翔;李吉泉;李治;孙涛. 基于XGBoost特征提取的数据驱动故障诊断方法[J]. 中国机械工程, 2020, 31(10): 1232-1239.
[12]	蔡英凤1;李健2;孙晓强1;陈龙1;江浩斌2;何友国1;陈小波1. 智能汽车路径跟踪混合控制策略研究[J]. 中国机械工程, 2020, 31(03): 289-298.
[13]	詹君1;程龙生1;彭宅铭1;胡多海2. 基于混合模型与改进多分类马田系统的控制图模式识别[J]. 中国机械工程, 2019, 30(22): 2716-2724.
[14]	刘朋;任工昌;杨力鹏;胡小龙. 基于斜率差的激光雷达环境特征提取方法[J]. 中国机械工程, 2019, 30(14): 1679-1683.
[15]	闫少文;屈文忠;肖黎. 一种基于高频频响函数的无基准疲劳裂纹识别方法[J]. 中国机械工程, 2019, 30(12): 1428-1432,1479.